Pixel-by-pixel correction of beam hardening artifacts by bowtie filter in fan-beam CT.

Objective. Beam hardening (BH) artifacts in computed tomography (CT) images originate from the polychromatic nature of x-ray photons. In a CT system with a bowtie filter, residual BH artifacts remain when polynomial fits are used. These artifacts lead to worse visuals, reduced contrast, and inaccurate CT numbers. This work proposes a pixel-by-pixel correction (PPC) method to reduce the residual BH artifacts caused by a bowtie filter.Approach. The energy spectrum for each pixel at the detector after the photons pass through the bowtie filter was calculated. Then, the spectrum was filtered through a series of water slabs with different thicknesses. The polychromatic projection corresponding to the thickness of the water slab for each detector pixel could be obtained. Next, we carried out a water slab experiment with a mono energyE= 69 keV to get the monochromatic projection. The polychromatic and monochromatic projections were then fitted with a 2nd-order polynomial. The proposed method was evaluated on digital phantoms in a virtual CT system and phantoms in a real CT machine.Main results. In the case of a virtual CT system, the standard deviation of the line profile was reduced by 23.8%, 37.3%, and 14.3%, respectively, in the water phantom with different shapes. The difference of the linear attenuation coefficients (LAC) in the central and peripheral areas of an image was reduced from 0.010 to 0.003cm-1and 0.007cm-1to 0 in the biological tissue phantom and human phantom, respectively. The method was also validated using CT projection data obtained from Activion16 (Canon Medical Systems, Japan). The difference in the LAC in the central and peripheral areas can be reduced by a factor of two.Significance. The proposed PPC method can successfully remove the cupping artifacts in both virtual and authentic CT images. The scanned object's shapes and materials do not affect the technique.

Feasibility of proton dosimetry overriding planning CT with daily CBCT elaborated through generative artificial intelligence tools.

Radiotherapy commonly utilizes cone beam computed tomography (CBCT) for patient positioning and treatment monitoring. CBCT is deemed to be secure for patients, making it suitable for the delivery of fractional doses. However, limitations such as a narrow field of view, beam hardening, scattered radiation artifacts, and variability in pixel intensity hinder the direct use of raw CBCT for dose recalculation during treatment. To address this issue, reliable correction techniques are necessary to remove artifacts and remap pixel intensity into Hounsfield Units (HU) values. This study proposes a deep-learning framework for calibrating CBCT images acquired with narrow field of view (FOV) systems and demonstrates its potential use in proton treatment planning updates. Cycle-consistent generative adversarial networks (cGAN) processes raw CBCT to reduce scatter and remap HU. Monte Carlo simulation is used to generate CBCT scans, enabling the possibility to focus solely on the algorithm's ability to reduce artifacts and cupping effects without considering intra-patient longitudinal variability and producing a fair comparison between planning CT (pCT) and calibrated CBCT dosimetry. To showcase the viability of the approach using real-world data, experiments were also conducted using real CBCT. Tests were performed on a publicly available dataset of 40 patients who received ablative radiation therapy for pancreatic cancer. The simulated CBCT calibration led to a difference in proton dosimetry of less than 2%, compared to the planning CT. The potential toxicity effect on the organs at risk decreased from about 50% (uncalibrated) up the 2% (calibrated). The gamma pass rate at 3%/2 mm produced an improvement of about 37% in replicating the prescribed dose before and after calibration (53.78% vs 90.26%). Real data also confirmed this with slightly inferior performances for the same criteria (65.36% vs 87.20%). These results may confirm that generative artificial intelligence brings the use of narrow FOV CBCT scans incrementally closer to clinical translation in proton therapy planning updates.

A 4D-CBCT correction network based on contrastive learning for dose calculation in lung cancer.

OBJECTIVE: This study aimed to present a deep-learning network called contrastive learning-based cycle generative adversarial networks (CLCGAN) to mitigate streak artifacts and correct the CT value in four-dimensional cone beam computed tomography (4D-CBCT) for dose calculation in lung cancer patients. METHODS: 4D-CBCT and 4D computed tomography (CT) of 20 patients with locally advanced non-small cell lung cancer were used to paired train the deep-learning model. The lung tumors were located in the right upper lobe, right lower lobe, left upper lobe, and left lower lobe, or in the mediastinum. Additionally, five patients to create 4D synthetic computed tomography (sCT) for test. Using the 4D-CT as the ground truth, the quality of the 4D-sCT images was evaluated by quantitative and qualitative assessment methods. The correction of CT values was evaluated holistically and locally. To further validate the accuracy of the dose calculations, we compared the dose distributions and calculations of 4D-CBCT and 4D-sCT with those of 4D-CT. RESULTS: The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) of the 4D-sCT increased from 87% and 22.31 dB to 98% and 29.15 dB, respectively. Compared with cycle consistent generative adversarial networks, CLCGAN enhanced SSIM and PSNR by 1.1% (p < 0.01) and 0.42% (p < 0.01). Furthermore, CLCGAN significantly decreased the absolute mean differences of CT value in lungs, bones, and soft tissues. The dose calculation results revealed a significant improvement in 4D-sCT compared to 4D-CBCT. CLCGAN was the most accurate in dose calculations for left lung (V5Gy), right lung (V5Gy), right lung (V20Gy), PTV (D98%), and spinal cord (D2%), with the relative dose difference were reduced by 6.84%, 3.84%, 1.46%, 0.86%, 3.32% compared to 4D-CBCT. CONCLUSIONS: Based on the satisfactory results obtained in terms of image quality, CT value measurement, it can be concluded that CLCGAN-based corrected 4D-CBCT can be utilized for dose calculation in lung cancer.

Feasibility of bone marrow edema detection using dual-energy cone-beam computed tomography.

BACKGROUND: Dual-energy (DE) detection of bone marrow edema (BME) would be a valuable new diagnostic capability for the emerging orthopedic cone-beam computed tomography (CBCT) systems. However, this imaging task is inherently challenging because of the narrow energy separation between water (edematous fluid) and fat (health yellow marrow), requiring precise artifact correction and dedicated material decomposition approaches. PURPOSE: We investigate the feasibility of BME assessment using kV-switching DE CBCT with a comprehensive CBCT artifact correction framework and a two-stage projection- and image-domain three-material decomposition algorithm. METHODS: DE CBCT projections of quantitative BME phantoms (water containers 100-165 mm in size with inserts presenting various degrees of edema) and an animal cadaver model of BME were acquired on a CBCT test bench emulating the standard wrist imaging configuration of a Multitom Rax twin robotic x-ray system. The slow kV-switching scan protocol involved a 60 kV low energy (LE) beam and a 120 kV high energy (HE) beam switched every 0.5° over a 200° angular span. The DE CBCT data preprocessing and artifact correction framework consisted of (i) projection interpolation onto matched LE and HE projections views, (ii) lag and glare deconvolutions, and (iii) efficient Monte Carlo (MC)-based scatter correction. Virtual non-calcium (VNCa) images for BME detection were then generated by projection-domain decomposition into an Aluminium (Al) and polyethylene basis set (to remove beam hardening) followed by three-material image-domain decomposition into water, Ca, and fat. Feasibility of BME detection was quantified in terms of VNCa image contrast and receiver operating characteristic (ROC) curves. Robustness to object size, position in the field of view (FOV) and beam collimation (varied 20-160 mm) was investigated. RESULTS: The MC-based scatter correction delivered > 69% reduction of cupping artifacts for moderate to wide collimations (> 80 mm beam width), which was essential to achieve accurate DE material decomposition. In a forearm-sized object, a 20% increase in water concentration (edema) of a trabecular bone-mimicking mixture presented as ∼15 HU VNCa contrast using 80-160 mm beam collimations. The variability with respect to object position in the FOV was modest (< 15% coefficient of variation). The areas under the ROC curve were > 0.9. A femur-sized object presented a somewhat more challenging task, resulting in increased sensitivity to object positioning at 160 mm collimation. In animal cadaver specimens, areas of VNCa enhancement consistent with BME were observed in DE CBCT images in regions of MRI-confirmed edema. CONCLUSION: Our results indicate that the proposed artifact correction and material decomposition pipeline can overcome the challenges of scatter and limited spectral separation to achieve relatively accurate and sensitive BME detection in DE CBCT. This study provides an important baseline for clinical translation of musculoskeletal DE CBCT to quantitative, point-of-care bone health assessment.

Improving the geographical origin classification of Radix glycyrrhizae (licorice) through hyperspectral imaging assisted by U-Net fine structure recognition.

Radix glycyrrhizae (licorice) is extensively employed in traditional Chinese medicine, and serves as a crucial raw material in industries such as food and cosmetics. The quality of licorice from different origins varies greatly, so classification of its geographical origin is particularly important. This study proposes a technique for fine structure recognition and segmentation of hyperspectral images of licorice using deep learning U-Net neural networks to segment the tissue structure patterns (phloem, xylem, and pith). Firstly, the three partitions were separately labeled using the Labelme tool, which was utilized to train the U-Net model. Secondly, the obtained optimal U-Net model was applied to predict three partitions of all samples. Lastly, various machine learning models (LDA, SVM, and PLS-DA) were trained based on segmented hyperspectral data. In addition, a threshold method and a circumcircle method were applied to segment licorice hyperspectral images for comparison. The results revealed that compared with the threshold segmentation method (which yielded SVM classifier accuracies of 99.17%, 91.15%, and 92.50% on the training set, validation set, and test set, respectively), the U-Net segmentation method significantly enhanced the accuracy of origin classification (99.06%, 94.72% and 96.07%). Conversely, the circumcircle segmentation method did not effectively improve the accuracy of origin classification (99.65%, 91.16% and 92.13%). By integrating Raman imaging of licorice, it can be inferred that the U-Net model, designed for region segmentation based on the inherent tissue structure of licorice, can effectively improve the accuracy origin classification, which has positive significance in the development of intelligence and information technology of Chinese medicine quality control.

[Deep learning approach for automatic segmentation of auricular acupoint divisions].

The automatic segmentation of auricular acupoint divisions is the basis for realizing intelligent auricular acupoint therapy. However, due to the large number of ear acupuncture areas and the lack of clear boundary, existing solutions face challenges in automatically segmenting auricular acupoints. Therefore, a fast and accurate automatic segmentation approach of auricular acupuncture divisions is needed. A deep learning-based approach for automatic segmentation of auricular acupoint divisions is proposed, which mainly includes three stages: ear contour detection, anatomical part segmentation and keypoints localization, and image post-processing. In the anatomical part segmentation and keypoints localization stages, K-YOLACT was proposed to improve operating efficiency. Experimental results showed that the proposed approach achieved automatic segmentation of 66 acupuncture points in the frontal image of the ear, and the segmentation effect was better than existing solutions. At the same time, the mean average precision (mAP) of the anatomical part segmentation of the K-YOLACT was 83.2%, mAP of keypoints localization was 98.1%, and the running speed was significantly improved. The implementation of this approach provides a reliable solution for the accurate segmentation of auricular point images, and provides strong technical support for the modern development of traditional Chinese medicine.

Segmenting hypothalamic subunits in human newborn magnetic resonance imaging data.

Preclinical evidence suggests that inter-individual variation in the structure of the hypothalamus at birth is associated with variation in the intrauterine environment, with downstream implications for future disease susceptibility. However, scientific advancement in humans is limited by a lack of validated methods for the automatic segmentation of the newborn hypothalamus. N = 215 healthy full-term infants with paired T1-/T2-weighted MR images across four sites were considered for primary analyses (mean postmenstrual age = 44.3 ± 3.5 weeks, nmale /nfemale = 110/106). The outputs of FreeSurfer's hypothalamic subunit segmentation tools designed for adults (segFS) were compared against those of a novel registration-based pipeline developed here (segATLAS) and against manually edited segmentations (segMAN) as reference. Comparisons were made using Dice Similarity Coefficients (DSCs) and through expected associations with postmenstrual age at scan. In addition, we aimed to demonstrate the validity of the segATLAS pipeline by testing for the stability of inter-individual variation in hypothalamic volume across the first year of life (n = 41 longitudinal datasets available). SegFS and segATLAS segmentations demonstrated a wide spread in agreement (mean DSC = 0.65 ± 0.14 SD; range = {0.03-0.80}). SegATLAS volumes were more highly correlated with postmenstrual age at scan than segFS volumes (n = 215 infants; RsegATLAS 2 = 65% vs. RsegFS 2 = 40%), and segATLAS volumes demonstrated a higher degree of agreement with segMAN reference segmentations at the whole hypothalamus (segATLAS DSC = 0.89 ± 0.06 SD; segFS DSC = 0.68 ± 0.14 SD) and subunit levels (segATLAS DSC = 0.80 ± 0.16 SD; segFS DSC = 0.40 ± 0.26 SD). In addition, segATLAS (but not segFS) volumes demonstrated stability from near birth to ~1 years age (n = 41; R2 = 25%; p < 10-3 ). These findings highlight segATLAS as a valid and publicly available (https://github.com/jerodras/neonate_hypothalamus_seg) pipeline for the segmentation of hypothalamic subunits using human newborn MRI up to 3 months of age collected at resolutions on the order of 1 mm isotropic. Because the hypothalamus is traditionally understudied due to a lack of high-quality segmentation tools during the early life period, and because the hypothalamus is of high biological relevance to human growth and development, this tool may stimulate developmental and clinical research by providing new insight into the unique role of the hypothalamus and its subunits in shaping trajectories of early life health and disease.

Accurate, automatic determination of astigmatism and phase with Ctfplotter in IMOD.

Ctfplotter in the IMOD software package is a flexible program for determination of CTF parameters in tilt series images. It uses a novel approach to find astigmatism by measuring defocus in one-dimensional power spectra rotationally averaged over a series of restricted angular ranges. Comparisons with Ctffind, Gctf, and Warp show that Ctfplotter's estimated astigmatism is generally more reliable than that found by these programs that fit CTF parameters to two-dimensional power spectra, especially at higher tilt angles. In addition to that intrinsic advantage, Ctfplotter can reduce the variability in astigmatism estimates further by summing results over multiple tilt angles (typically 5), while still finding defocus for each individual image. Its fitting strategy also produces better phase estimates. The program now includes features for tuning the sampling of the power spectrum so that it is well-represented for analysis, and for determining an appropriate fitting range that can vary with tilt angle. It can thus be used automatically in a variety of situations, not just for fitting tilt series, and has been integrated into the SerialEM acquisition software for real-time determination of focus and astigmatism.

An Automatic Method for Sublingual Image Segmentation and Color Analysis.

Automatic segmentation of sublingual images and color quantification of sublingual vein are of great significance to disease diagnosis in traditional Chinese medicine. With the development of computer vision, automatic sublingual image processing provides a noninvasive way to observe patients' tongue and is convenient for both doctors and patients. However, current sublingual image segmentation methods are not accurate enough. Besides, the differences in subjective judgments by different doctors bring more difficulties in color analysis of sublingual veins. In this paper, we propose a method of sublingual image segmentation based on a modified UNet++ network to improve the segmentation accuracy, a color classification approach based on triplet network, and a color quantization method of sublingual vein based on linear discriminant analysis to provide intuitive one-dimensional results. Our methods achieve 88.2% mean intersection over union (mIoU) and 94.1% pixel accuracy on tongue dorsum segmentation, and achieves 69.8% mIoU and 82.7% pixel accuracy on sublingual vein segmentation. Compared with the state-of-art methods, the segmentation mIoUs are improved by 5.8% and 5.3% respectively. Our sublingual vein color classification method has the highest overall accuracy of 81.2% and the highest recall for the minority class of 77.5%, and the accuracy of color quantization is 90.5%.Clinical Relevance- The methods provide accurate and quantified information of the sublingual image, which can assist doctors in diagnosis.

Image quality improvement in bowtie-filter-equipped cone-beam CT using a dual-domain neural network.

BACKGROUND: The bowtie-filter in cone-beam CT (CBCT) causes spatially nonuniform x-ray beam often leading to eclipse artifacts in the reconstructed image. The artifacts are further confounded by the patient scatter, which is therefore patient-dependent as well as system-specific. PURPOSE: In this study, we propose a dual-domain network for reducing the bowtie-filter-induced artifacts in CBCT images. METHODS: In the projection domain, the network compensates for the filter-induced beam-hardening that are highly related to the eclipse artifacts. The output of the projection-domain network was used for image reconstruction and the reconstructed images were fed into the image-domain network. In the image domain, the network further reduces the remaining cupping artifacts that are associated with the scatter. A single image-domain-only network was also implemented for comparison. RESULTS: The proposed approach successfully enhanced soft-tissue contrast with much-reduced image artifacts. In the numerical study, the proposed method decreased perceptual loss and root-mean-square-error (RMSE) of the images by 84.5% and 84.9%, respectively, and increased the structure similarity index measure (SSIM) by 0.26 compared to the original input images on average. In the experimental study, the proposed method decreased perceptual loss and RMSE of the images by 87.2% and 92.1%, respectively, and increased SSIM by 0.58 compared to the original input images on average. CONCLUSIONS: We have proposed a deep-learning-based dual-domain framework to reduce the bowtie-filter artifacts and to increase the soft-tissue contrast in CBCT images. The performance of the proposed method has been successfully demonstrated in both numerical and experimental studies.

Ultrasound image segmentation using Gamma combined with Bayesian model for focused-ultrasound-surgery lesion recognition.

This study aims to investigate the feasibility of combined segmentation for the separation of lesions from non-ablated regions, which allows surgeons to easily distinguish, measure, and evaluate the lesion area, thereby improving the quality of high-intensity focused-ultrasound (HIFU) surgery used for the non-invasive tumor treatment. Given that the flexible shape of the Gamma mixture model (GΓMM) fits the complex statistical distribution of samples, a method combining the GΓMM and Bayes framework is constructed for the classification of samples to obtain the segmentation result. An appropriate normalization range and parameters can be used to rapidly obtain a good performance of GΓMM segmentation. The performance values of the proposed method under four metrics (Dice score: 85%, Jaccard coefficient: 75%, recall: 86%, and accuracy: 96%) are better than those of conventional approaches including Otsu and Region growing. Furthermore, the statistical result of sample intensity indicates that the finding of the GΓMM is similar to that obtained by the manual method. These results indicate the stability and reliability of the GΓMM combined with the Bayes framework for the segmentation of HIFU lesions in ultrasound images. The experimental results show the possibility of combining the GΓMM with the Bayes framework to segment lesion areas and evaluate the effect of therapeutic ultrasound.

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application.

Tongue diagnosis is an essential technique of traditional Chinese medicine (TCM) diagnosis, and the need for objectifying tongue images through image processing technology is growing. The present study provides an overview of the progress made in tongue objectification over the past decade and compares segmentation models. Various deep learning models are constructed to verify and compare algorithms using real tongue image sets. The strengths and weaknesses of each model are analyzed. The findings indicate that the U-Net algorithm outperforms other models regarding precision accuracy (PA), recall, and mean intersection over union (MIoU) metrics. However, despite the significant progress in tongue image acquisition and processing, a uniform standard for objectifying tongue diagnosis has yet to be established. To facilitate the widespread application of tongue images captured using mobile devices in tongue diagnosis objectification, further research could address the challenges posed by tongue images captured in complex environments.

Generative adversarial networks in EEG analysis: an overview.

Electroencephalogram (EEG) signals have been utilized in a variety of medical as well as engineering applications. However, one of the challenges associated with recording EEG data is the difficulty of recording large amounts of data. Consequently, data augmentation is a potential solution to overcome this challenge in which the objective is to increase the amount of data. Inspired by the success of Generative Adversarial Networks (GANs) in image processing applications, generating artificial EEG data from the limited recorded data using GANs has seen recent success. This article provides an overview of various techniques and approaches of GANs for augmenting EEG signals. We focus on the utility of GANs in different applications including Brain-Computer Interface (BCI) paradigms such as motor imagery and P300-based systems, in addition to emotion recognition, epileptic seizures detection and prediction, and various other applications. We address in this article how GANs have been used in each study, the impact of using GANs on the model performance, the limitations of each algorithm, and future possibilities for developing new algorithms. We emphasize the utility of GANs in augmenting the limited EEG data typically available in the studied applications.

Evaluation of a two-dimensional Moire-free antiscatter grid for cone-beam computed tomography.

BACKGROUND: Scatter radiation is a traditional problem in cone-beam computed tomography (CBCT). Accordingly, numerous methods have been researched for scatter reduction in terms of software- or hardware-based solutions. The concept of a two-dimensional antiscatter grid (2D-ASG) has been shown to provide a solution to the scatter reduction in CBCT. However, the use of an ASG makes it challenging to use in clinical CBCT systems because it causes Moire artifacts of the image. PURPOSE: We have developed a Moire-free 2D-ASG that was designed to solve the Moire artifact and the scatter radiation problems. We provide the experimental results pertaining to the image quality measurements from the evaluation of the 2D-ASG compared with those obtained from a one-dimensional antiscatter grid (1D-ASG) and no antiscatter grid (No-ASG) to demonstrate the quantitative extent of the improvements. METHODS: The 2D-ASG, fabricated based on a sawing process with a graphite body, was prepared to evaluate image quality improvements. Projection images for Pro-CT MK II phantom were acquired using the CBCT testbed of sample rotation type and reconstructed by Feldkamp-Davis-Kress (FDK) algorithm without any filters. We measured the contrast-to-noise ratio (CNR) and uniformity index for the cupping artifacts of the 2D-ASG, 1D-ASG, and No-ASG cases. RESULTS: 2D-ASG considerably reduced the cupping artifacts owing to the scatter reduction compared with the 1D-ASG and No-ASG cases. The cupping artifacts were reduced by 85% in the 2D-ASG compared with No-ASG, while the cupping artifacts were reduced by 63% in 1D-ASG compared with No-ASG. The 2D-ASG also yielded a CNR improvement. The average CNR improvements for eight insert materials were 47% in the 2D-ASG compared with No-ASG, while the CNR was improved by 36% in the 1D-ASG compared with No-ASG. CONCLUSIONS: We demonstrated that the Moire-free 2D-ASG improved the image quality by removing scatter radiation in CBCT compared with 1D-ASG and No-ASG. We believe that the Moire-free 2D-ASG can become one of the effective ways to solve the scatter radiation problem in CBCT images because it provides usability and has the potential to have synergistic effects on other methods, such as bow-tie filter and scatter correction.

Automatic segmentation of the thalamus using a massively trained 3D convolutional neural network: higher sensitivity for the detection of reduced thalamus volume by improved inter-scanner stability.

OBJECTIVES: To develop an automatic method for accurate and robust thalamus segmentation in T1w-MRI for widespread clinical use without the need for strict harmonization of acquisition protocols and/or scanner-specific normal databases. METHODS: A three-dimensional convolutional neural network (3D-CNN) was trained on 1975 T1w volumes from 170 MRI scanners using thalamus masks generated with FSL-FIRST as ground truth. Accuracy was evaluated with 18 manually labeled expert masks. Intra- and inter-scanner test-retest stability were assessed with 477 T1w volumes of a single healthy subject scanned on 123 MRI scanners. The sensitivity of 3D-CNN-based volume estimates for the detection of thalamus atrophy was tested with 127 multiple sclerosis (MS) patients and a normal database comprising 4872 T1w volumes from 160 scanners. The 3D-CNN was compared with a publicly available 2D-CNN (FastSurfer) and FSL. RESULTS: The Dice similarity coefficient of the automatic thalamus segmentation with manual expert delineation was similar for all tested methods (3D-CNN and FastSurfer 0.86 ± 0.02, FSL 0.87 ± 0.02). The standard deviation of the single healthy subject's thalamus volume estimates was lowest with 3D-CNN for repeat scans on the same MRI scanner (0.08 mL, FastSurfer 0.09 mL, FSL 0.15 mL) and for repeat scans on different scanners (0.28 mL, FastSurfer 0.62 mL, FSL 0.63 mL). The proportion of MS patients with significantly reduced thalamus volume was highest for 3D-CNN (24%, FastSurfer 16%, FSL 11%). CONCLUSION: The novel 3D-CNN allows accurate thalamus segmentation, similar to state-of-the-art methods, with considerably improved robustness with respect to scanner-related variability of image characteristics. This might result in higher sensitivity for the detection of disease-related thalamus atrophy. KEY POINTS: • A three-dimensional convolutional neural network was trained for automatic segmentation of the thalamus with a heterogeneous sample of T1w-MRI from 1975 patients scanned on 170 different scanners. • The network provided high accuracy for thalamus segmentation with manual segmentation by experts as ground truth. • Inter-scanner variability of thalamus volume estimates across different MRI scanners was reduced by more than 50%, resulting in increased sensitivity for the detection of thalamus atrophy.

3D-printed large-area focused grid for scatter reduction in cone-beam CT.

BACKGROUND: Cone-beam computed tomography (CBCT) systems acquire volumetric data more efficiently than fan-beam or multislice CT, particularly when the anatomy of interest resides within the axial field-of-view of the detector and data can be acquired in one rotation. For such systems, scattered radiation remains a source of image quality degradation leading to increased noise, image artifacts, and CT number inaccuracies. PURPOSE: Recent advances in metal additive manufacturing allow the production of highly focused antiscatter grids (2D-ASGs) that can be used to reduce scatter intensity, while preserving primary radiation transmission. We present the first implementation of a large-area, 2D-ASG for flat-panel CBCT, including grid-line artifact removal and related improvements in image quality. METHODS: A 245 × 194 × 10 mm 2D-ASG was manufactured from chrome-cobalt alloy using laser powder-bed fusion (LPBF) (AM-400; Renishaw plc, New Mills Wotton-under-Edge, UK). The 2D-ASG had a square profile with a pitch of 9.09 lines/cm and 10:1 grid-ratio. The nominal 0.1 mm grid septa were focused to a 732 mm x-ray source to optimize primary x-ray transmission and reduce grid-line shadowing at the detector. Powder-bed fusion ensured the structural stability of the ASG with no need for additional interseptal support. The 2D-ASG was coupled to a 0.139-mm element pitch flat-panel detector (DRX 3543, Carestream Health) and proper alignment was confirmed by consistent grid-line shadow thickness across the whole detector array. A 154-mm diameter CBCT image-quality-assurance phantom was imaged using a rotary stage and a ceiling-mounted, x-ray unit (Proteus XR/a, GE Medical Systems, 80kVp, 0.5mAs). Grid-line artifacts were removed using a combination of exposure-dependent gain correction and spatial-frequency, Fourier filtering. Projections were reconstructed using a Parker-weighted, FDK algorithm and voxels were spatially averaged to 357 × 357 × 595 µm to improve the signal-to-noise characteristics of the CBCT reconstruction. Finally, in order to compare image quality with and without scatter, the phantom was scanned again under the same CBCT conditions but with no 2D-ASG. No additional antiscatter (i.e., air-gap, bowtie filtration) strategies were used to evaluate the effects in image quality caused by the 2D-ASG alone. RESULTS: The large-area, 2D-ASG prototype was successfully designed and manufactured using LPBF. CBCT image-quality improvements using the 2D-ASG included: an overall 14.5% CNR increase across the volume; up to 48.8% CNR increase for low-contrast inserts inside the contrast plate of the QA phantom; and a 65% reduction of cupping artifact in axial profiles of water-filled cross sections of the phantom. Advanced image processing strategies to remove grid line artifacts did not affect the spatial resolution or geometric accuracy of the system. CONCLUSIONS: LPBF can be used to manufacture highly efficient, 2D-focused ASGs that can be easily coupled to clinical, flat-panel detectors. The implementation of ASGs in CBCT leads to reduced scatter-related artifacts, improved CT number accuracy, and enhanced CNR with no increased equivalent dose to the patient. Further improvements to image quality might be achieved with a combination of scatter-correction algorithms and iterative-reconstruction strategies. Finally, clinical applications where other scatter removal strategies are unfeasible might now achieve superior soft-tissue visualization and quantitative capabilities.

Learning from dermoscopic images in association with clinical metadata for skin lesion segmentation and classification.

Automatic segmentation and classification of lesions are two clinically significant tasks in the computer-aided diagnosis of skin diseases. Both tasks are challenging due to the nonnegligible lesion differences in dermoscopic images from different patients. In this paper, we propose a novel pipeline to efficiently implement skin lesions' segmentation and classification tasks, which consists of a segmentation network and a classification network. To improve the performance of the segmentation network, we propose a novel module of Multi-Scale Holistic Feature Exploration (MSH) to thoroughly exploit perceptual clues latent among multi-scale feature maps as synthesized by the decoder. The MSH module enables holistic exploration of features across multiple scales to more effectively support downstream image analysis tasks. To boost the performance of the classification network, we propose a novel module of Cross-Modality Collaborative Feature Exploration (CMC) to discover latent discriminative features by collaboratively exploiting potential relationships between cross-modal features of dermoscopic images and clinical metadata. The CMC module enables dynamically capturing versatile interaction effects among cross-modal features during the model's representation learning procedure by discriminatively and adaptively learning the interaction weight associated with each crossmodality feature pair. In addition, to effectively reduce background noise and boost the lesion discrimination ability of the classification network, we crop the images based on lesion masks generated by the best segmentation model. We evaluate the proposed pipeline on the four public skin lesion datasets, where the ISIC 2018 and PH2 are for segmentation, and the ISIC 2019 and ISIC 2020 are combined into a new dataset, ISIC 2019&2020, for classification. It achieves a Jaccard index of 83.31% and 90.14% in skin lesion segmentation, an AUC of 97.98% and an Accuracy of 92.63% in skin lesion classification, which is superior to the performance of representative state-of-the-art skin lesion segmentation and classification methods. Last but not least, the new model for segmentation utilizes much fewer model parameters (3.3 M) than its peer approaches, leading to a greatly reduced number of labeled samples required for model training, which obtains substantially stronger robustness than its peers.

Deep learning for x-ray scatter correction in dedicated breast CT.

BACKGROUND: Accurate correction of x-ray scatter in dedicated breast computed tomography (bCT) imaging may result in improved visual interpretation and is crucial to achieve quantitative accuracy during image reconstruction and analysis. PURPOSE: To develop a deep learning (DL) model to correct for x-ray scatter in bCT projection images. METHODS: A total of 115 patient scans acquired with a bCT clinical system were segmented into the major breast tissue types (skin, adipose, and fibroglandular tissue). The resulting breast phantoms were divided into training (n = 110) and internal validation cohort (n = 5). Training phantoms were augmented by a factor of four by random translation of the breast in the image field of view. Using a previously validated Monte Carlo (MC) simulation algorithm, 12 primary and scatter bCT projection images with a 30-degree step were generated from each phantom. For each projection, the thickness map and breast location in the field of view were also calculated. A U-Net based DL model was developed to estimate the scatter signal based on the total input simulated image and trained single-projection-wise, with the thickness map and breast location provided as additional inputs. The model was internally validated using MC-simulated projections and tested using an external data set of 10 phantoms derived from images acquired with a different bCT system. For this purpose, the mean relative difference (MRD) and mean absolute error (MAE) were calculated. To test for accuracy in reconstructed images, a full bCT acquisition was mimicked with MC-simulations and then assessed by calculating the MAE and the structural similarity (SSIM). Subsequently, scatter was estimated and subtracted from the bCT scans of three patients to obtain the scatter-corrected image. The scatter-corrected projections were reconstructed and compared with the uncorrected reconstructions by evaluating the correction of the cupping artifact, increase in image contrast, and contrast-to-noise ratio (CNR). RESULTS: The mean MRD and MAE across all cases (min, max) for the internal validation set were 0.04% (-1.1%, 1.3%) and 2.94% (2.7%, 3.2%), while for the external test set they were -0.64% (-1.6%, 0.2%) and 2.84% (2.3%, 3.5%), respectively. For MC-simulated reconstruction slices, the computed SSIM was 0.99 and the MAE was 0.11% (range: 0%, 0.35%) with a single outlier slice of 2.06%. For the three patient bCT reconstructed images, the correction increased the contrast by a mean of 25% (range: 20%, 30%), and reduced the cupping artifact. The mean CNR increased by 0.32 after scatter correction, which was not found to be significant (95% confidence interval: [-0.01, 0.65], p = 0.059). The time required to correct the scatter in a single bCT projection was 0.2 s on an NVIDIA GeForce GTX 1080 GPU. CONCLUSION: The developed DL model could accurately estimate scatter in bCT projection images and could enhance contrast and correct for cupping artifact in reconstructed patient images without significantly affecting the CNR. The time required for correction would allow its use in daily clinical practice, and the reported accuracy will potentially allow quantitative reconstructions.

Multi-Object Detection in Security Screening Scene Based on Convolutional Neural Network.

The technique for target detection based on a convolutional neural network has been widely implemented in the industry. However, the detection accuracy of X-ray images in security screening scenarios still requires improvement. This paper proposes a coupled multi-scale feature extraction and multi-scale attention architecture. We integrate this architecture into the Single Shot MultiBox Detector (SSD) algorithm and find that it can significantly improve the effectiveness of target detection. Firstly, ResNet is used as the backbone network to replace the original VGG network to improve the feature extraction capability of the convolutional neural network for images. Secondly, a multi-scale feature extraction (MSE) structure is designed to enrich the information contained in the multi-stage prediction feature layer. Finally, the multi-scale attention architecture (MSA) is fused onto the prediction feature layer to eliminate the redundant features' interference and extract effective contextual information. In addition, a combination of Adaptive-NMS and Soft-NMS is used to output the final prediction anchor boxes when performing non-maximum suppression. The results of the experiments show that the improved method improves the mean average precision (mAP) value by 7.4% compared to the original approach. New modules make detection much more accurate while keeping the detection speed the same.

Subcortical and hippocampal brain segmentation in 5-year-old children: Validation of FSL-FIRST and FreeSurfer against manual segmentation.

Developing accurate subcortical volumetric quantification tools is crucial for neurodevelopmental studies, as they could reduce the need for challenging and time-consuming manual segmentation. In this study, the accuracy of two automated segmentation tools, FSL-FIRST (with three different boundary correction settings) and FreeSurfer, were compared against manual segmentation of the hippocampus and subcortical nuclei, including the amygdala, thalamus, putamen, globus pallidus, caudate and nucleus accumbens, using volumetric and correlation analyses in 80 5-year-olds. Both FSL-FIRST and FreeSurfer overestimated the volume on all structures except the caudate, and the accuracy varied depending on the structure. Small structures such as the amygdala and nucleus accumbens, which are visually difficult to distinguish, produced significant overestimations and weaker correlations with all automated methods. Larger and more readily distinguishable structures such as the caudate and putamen produced notably lower overestimations and stronger correlations. Overall, the segmentations performed by FSL-FIRST's default pipeline were the most accurate, whereas FreeSurfer's results were weaker across the structures. In line with prior studies, the accuracy of automated segmentation tools was imperfect with respect to manually defined structures. However, apart from amygdala and nucleus accumbens, FSL-FIRST's agreement could be considered satisfactory (Pearson correlation > 0.74, intraclass correlation coefficient (ICC) > 0.68 and Dice score coefficient (DSC) > 0.87) with highest values for the striatal structures (putamen, globus pallidus, caudate) (Pearson correlation > 0.77, ICC > 0.87 and DSC > 0.88, respectively). Overall, automated segmentation tools do not always provide satisfactory results, and careful visual inspection of the automated segmentations is strongly advised.